Is this a sensor problem or a lens problem or my skills?

[boombox]

Bro, If you're caught without a bounce card or an omnibounce - If there's a pull out wide angle diffuser, you can pull it almost all the way out - so it sticks out vertically - doubling up as a tiny bounce card, when the ceiling is too high, angle the flash head forwards with the "mini" bounce card on.

 

[TME]

Bro, If you're caught without a bounce card or an omnibounce - If there's a pull out wide angle diffuser, you can pull it almost all the way out - so it sticks out vertically - doubling up as a tiny bounce card, when the ceiling is too high, angle the flash head forwards with the "mini" bounce card on.

Mine doesn't stand up properly leh... also it is transparent, does it reflect enough light? I know the Nikon flashes have a proper bounce card attached...

 

[UY79]

Well this is definitely something that you've got to help me on cos I'm quite unclear about it... what's the difference between f/1.4 focussed at infinity and f/8 focussed at infinity... assuming I focussed on the people in red in my posted photo, does that mean that the DOF will be measured at the people in red? So if I shoot at a very shallow DOF, then it is easy to get these people OOF? Does it work this way? Usually I just adjust to f/8 but I never really got it where focus at infinity is concerned. Anyone kind enough to enlighten me?

You have some serious misconceptions about focus and DOF.

DOF falls ahead of and after the point of focus. So when you put the focus at infinity, you will have DOF that starts at infinity. This also meant that your subjects can never be in the DOF as the DOF is at an infinity distance away.

In your case, your whole photo was not in focus because no part of it was in DOF if you focused at infinity or past infinity.

Try reading this too: http://en.wikipedia.org/wiki/Depth_of_field.

 

[TME]

You have some serious misconceptions about focus and DOF.

DOF falls ahead of and after the point of focus. So when you put the focus at infinity, you will have DOF that starts at infinity. This also meant that your subjects can never be in the DOF as the DOF is at an infinity distance away.

In your case, your whole photo was not in focus because no part of it was in DOF if you focused at infinity or past infinity.

Try reading this too: http://en.wikipedia.org/wiki/Depth_of_field.

Hmm... I focused specifically at the people in red (their faces from what I can remember) in the distances at f/4. So while the DOF is shallow, the trees, people in red and fence should ostensibly fall within the DOF right? When I say focus at infinity, it's because the scale on my lens read infinity when focus was confirmed by the camera's AF module. So I assumed that means the focus was at infinity? Is that right? I did not shoot in manual focus mode where you can deliberately set the focus of the lens to infinity and shoot. I gather that some people do that in certain situations where the focus point is far away and does not change...

 

[TME]

I'v tried to replicate the way I shoot the above picture. However as it is late, I had to make do with a tripod shoot...

My settings:

JPEG X Fine, ISO200, tripod, cable release, AF-S, aperture priority, centre-weighted metering. Lens used is a KM 17-35mm f/2.8-4. Camera is a 7D. No flash. Central focus sensor used. Point of focus is dead centre of picture.

I shot a series from f/2.8 to f/11. In each case, the AF module placed the scale at infinity.

This first picture is the original. The exposure is the same for all the four shots. The nearest edge of the building is about 50m from my window, the central lift shaft (blue) is about 100m or more away.

I post 100% crops (centre) for each of the apertures I shot at. They all seem just as sharp to me except for the first one @ f/2.8 which seems to be optical flaw rather than OOF...

Full picture of scene (f/2.8)

100% crop @ f/2.8

100% crop @ f/5.6

100% crop @ f/8

100% crop @ f/11

I just want to ask if this test makes any sense... what I am seeing is that the DOF is quite large and as the building is slanted to the sensor plane, quite a bit of the image is in focus still, even at f/2.8. Hence for the flat frontal shot in China posted earlier, I am quite curious as to why the focus is so off... the distance from where I stood for the coal mine picture is roughly about 200-250m away. But it's quite clear that the background is very much softer than what I got in my test shots from f/2.8 to f/11. So then is it a resolution issue or my simply poor understanding of DOF, which I must admit isn't very good I think???

 

[UY79]

Hmm... I focused specifically at the people in red (their faces from what I can remember) in the distances at f/4. So while the DOF is shallow, the trees, people in red and fence should ostensibly fall within the DOF right? When I say focus at infinity, it's because the scale on my lens read infinity when focus was confirmed by the camera's AF module. So I assumed that means the focus was at infinity? Is that right? I did not shoot in manual focus mode where you can deliberately set the focus of the lens to infinity and shoot. I gather that some people do that in certain situations where the focus point is far away and does not change...

I've no idea where you got the notion that if the camera that put the lense's focus distance scale "at" the infinity mark that your camera's focus is indeed at infinity.

Have you ever have a good look at the distance representation on the scale? It's either logarithmic or exponential function that correspond to the turn of the focus ring. Usually beyond 10m, it just takes a very slight turn for the lense to focus to infinity. This means that if the camera focuses at a subject 100m away, the distance reading will be very close to infinity. But it's NOT at infinity.

I recommend that you do not manual focus the lense to infinity and follow whoever that suggested that it's a good idea. As I've explained earlier, focusing at infinity means that you put the whole image outside DOF. It will be made worse if your lense can focus past infinity.

If you want the whole image to be sharp in landscape photography, you should go learn about hyperfocal instead. http://en.wikipedia.org/wiki/Hyperfocal_distance

 

[TME]

The point is that I did not put the lens at infinity manually... the camera was on autofocus, wherever the scale ended was dependent on the AF module... in other words, based on whatever UY79 said in the previous posts, what he says doesn't hold true since I did not focus at infinity manually and did not put the whole scene out of the DOF... I'm merely assuming (wrongly) that the lens was focused at infinity since it is extremely close to the infinity mark on the scale and I am not clear as to the precision of the marking... I merely allowed the camera to do what it's supposed to do... and therefore I am questioning why the image is not as sharp as I expect it to be... I've tried to replicate my shooting technique... and on thinking back, I agree that the lens' scale was not at infinity... I was not aware that the scale is a logarithmic one... in which case the lens is definitely not focused at infinity but at some distant point... in the series that I shot, you can see that the images are sharp enough given the quality of the lens and the size of the sensor... but in the much brighter scene in China, the sharpness is pretty poor... so my question still stands... what's the problem?... Although I don't really understand how to set hyperfocal distance, from what I am aware of, I don't suppose that if I do not shoot at the hyperfocal distance, the image should be so unsharp (particularly at the centre of the image close to the focus point)... I would expect that shooting at the hyperfocal distance should yield the sharpest possible image (correct me if I am wrong)... the problem is I never understood how to set hyperfocal distance since all references that I have read so far have simply explained what it is and never gave an example of how to do so... including the Wikipedia reference... having a list of formidable equations is pretty useless since I would not be calculating these in the field... I would appreciate it if anyone would be kind enough to either reference something that I can read and actually practise or at least explain how to go about it...

 

[TME]

Ok, I've goggled a bit for some practical help and found something... it seems that you need a DOF scale on the camera... most of my lenses do not carry this scale except for the 50mm f/1.7 that I own... so how do you set hperfocal distance with lenses that don't carry a DOF scale?

This is helpful...
http://www.great-landscape-photography.com/hyperfocal.html

 

[Kit]

You need a hyperfocal chart...
http://www.clubsnap.com/forums/showthread.php?t=149057

 

[tunster]

the first shot looks very sharpened to me. DId you apply any effects on your camera or PP for that?

The second one is because of the underexposed of the left person. His faced captured more ambient lights so it turns red (It should be a room in yellow lighting). What you can do is to use a more spreading light diffuser (Maybe a light sphere or an omni bounce).

 

[TME]

I see... found another link that I found helpful as well...

http://www.dofmaster.com/hyperfocal.html

 

[TME]

the first shot looks very sharpened to me. DId you apply any effects on your camera or PP for that?

The second one is because of the underexposed of the left person. His faced captured more ambient lights so it turns red (It should be a room in yellow lighting). What you can do is to use a more spreading light diffuser (Maybe a light sphere or an omni bounce).

Ah, thanks... for the first shot, I think I did not sharpen much at all... I can't really remember... but if I did, I usually just use 100, 1.0, 0 in Photoshop.

The second shot, thanks for the advice... will try again when the opportunity arises...

 

[Crappinni]

Hi there.

The first photo (the one taken in China) seems like it was oversharpened from a soft image.

Perhaps you could check your camera settings to make sure there's no sharpening so you can see a more accurate representation of the problem?

 

[TME]

Hi there.

The first photo (the one taken in China) seems like it was oversharpened from a soft image.

Perhaps you could check your camera settings to make sure there's no sharpening so you can see a more accurate representation of the problem?

I probably sharpened the China photo and that's probably because it was pretty soft... which is my question in the first place... why is it soft? I notice this issue with my other general landscape shots... when viewed at 100%, the objects in the distance are soft... unless it's lacking in detail which gives the appearence of softness?

I leave my camera on +1 sharpening... it's very mild, leaves headroom for sharpening even in JPEG and reduces the need to sharpen in PS at all if I'm not too particular or if I'm sending to a photo lab for snap shots...

 

[UY79]

Since you said that you have used the camera AF on the subjects which is about 100m away, I suspect that the lack of sharpness in your China photo could be due to UV radiation.

What is the sea level elevation where you took the photo? And what filter did you have on the lense?

 

[TME]

Since you said that you have used the camera AF on the subjects which is about 100m away, I suspect that the lack of sharpness in your China photo could be due to UV radiation.

What is the sea level elevation where you took the photo? And what filter did you have on the lense?

Oh UV radiation? That's new to me? Could you explain its effects on focusing?

I was at Ningxia, it's a desert region, in Inner Mongolia, I have no idea if it was elevated, can't find anything on the net either. So being a desert region, it's cloudness almost all the time, when I was there, the region was in a 3-year drought period. So yes, I did get lots of UV radiation, I was very tanned when I got back even though we were teaching in a school most of the time... the sunlight was extremely harsh... the time of the day when that picture was taken was about 4 - 5 pm, the sun usually set at around.

Only filter on the camera was a Hoya SHMC UV filter. Same filter on both my Tamron 24-135mm SP and KM 17-35mm lenses.

 

[UY79]

Oh UV radiation? That's new to me? Could you explain its effects on focusing?

I was at Ningxia, it's a desert region, in Inner Mongolia, I have no idea if it was elevated, can't find anything on the net either. So being a desert region, it's cloudness almost all the time, when I was there, the region was in a 3-year drought period. So yes, I did get lots of UV radiation, I was very tanned when I got back even though we were teaching in a school most of the time... the sunlight was extremely harsh... the time of the day when that picture was taken was about 4 - 5 pm, the sun usually set at around.

Only filter on the camera was a Hoya SHMC UV filter. Same filter on both my Tamron 24-135mm SP and KM 17-35mm lenses.

Through my searches, Ningxia is 2000m above sea level. UV should be about 30% more than at sea level.

I can't really say for certain UV radiation that was the cause. But generally if you got photos that look like the ones you have shown in Ningxia but not had any back in Singapore, it could be it.

Some links:
http://www.cs.mtu.edu/~shene/DigiCam/User-Guide/filter/filter-UV.html
http://photo.net/equipment/filters/

 

[TME]

Thanks for the links... I've read the second one before... I'm not sure that UV radiation would cause such a significant deterioration of sharpness... I thought it causes hazing and hence you get this whitish film over the entire picture causing the tonal transitions to lose accutance leading to a perceived loss of sharpness or rendering the picture unsharp... that's what I understand the effects of UV radiation to do...

I posted this same problem over on dpreview... there's one person who said he gets this problem intermittently and it seems to usually vegetation framed by a clear or blue sky... in my case, it's usually vegetation but the attendant objects (people, buildings, animals, etc) also lose their sharpness... it becomes like blotchy... the China photo is a little exaggerated maybe because of over-enthusiastic sharpening... but that's the kind of pictures I sometimes get... often enough to trouble me cos some shots I don't get a chance to reshoot... or I should most shots I don't get a chance to reshoot...:dunno:

 

[UY79]

Thanks for the links... I've read the second one before... I'm not sure that UV radiation would cause such a significant deterioration of sharpness... I thought it causes hazing and hence you get this whitish film over the entire picture causing the tonal transitions to lose accutance leading to a perceived loss of sharpness or rendering the picture unsharp... that's what I understand the effects of UV radiation to do...

You are contradicting with yourself here. :think:

 

[Rashkae]

Through my searches, Ningxia is 2000m above sea level. UV should be about 30% more than at sea level.

I can't really say for certain UV radiation that was the cause. But generally if you got photos that look like the ones you have shown in Ningxia but not had any back in Singapore, it could be it.

Ehhh... All modern lenses already have UV filters as part of the glass and coating. A UV filter will have no effect. And even at 5900 meters my camera had no focusing problems. So I think the "UV Radiation theory" can be tossed out.